• Publications
  • Influence
Adaptive Control Processes
TLDR
This is a kind of book that you need now. Expand
  • 354
  • 25
Multi-task reinforcement learning: a hierarchical Bayesian approach
TLDR
We consider the problem of multi-task reinforcement learning, where the agent needs to solve a sequence of Markov Decision Processes (MDPs) chosen randomly from a fixed but unknown distribution. Expand
  • 230
  • 17
  • PDF
Dynamic preferences in multi-criteria reinforcement learning
TLDR
In this paper, we consider the problem of learning in the presence of time-varying preferences among multiple objectives, using numeric weights to represent their importance. Expand
  • 101
  • 10
  • PDF
Lower Bounding Klondike Solitaire with Monte-Carlo Planning
TLDR
We study Klondike using several sampling based planning approaches including UCT, hindsight optimization, and sparse sampling, and establish empirical lower bounds on their performance. Expand
  • 88
  • 10
  • PDF
Active Learning with Committees for Text Categorization
TLDR
We propose an active learning method that uses a committee of learners to reduce the number of training examples required for learning by l-2 orders of magnitude. Expand
  • 186
  • 9
  • PDF
Automatic discovery and transfer of MAXQ hierarchies
TLDR
We present an algorithm, HI-MAT (Hierarchy Induction via Models And Trajectories), that discovers MAXQ task hierarchies by applying dynamic Bayesian network models to a successful trajectory from a source reinforcement learning task. Expand
  • 85
  • 9
  • PDF
Learning First-Order Acyclic Horn Programs from Entailment
TLDR
We show that any subclass of first-order acyclic Horn programs with constant arity is exactly learnable from equivalence and entailment membership queries. Expand
  • 43
  • 9
A Decision-Theoretic Model of Assistance
TLDR
We formulate the problem of intelligent assistance in a decision-theoretic framework, and present both theoretical and empirical results. Expand
  • 92
  • 8
  • PDF
Using trajectory data to improve bayesian optimization for reinforcement learning
TLDR
We show how to more effectively apply Bayesian Optimization to RL by exploiting the sequential trajectory information generated by RL agents. Expand
  • 55
  • 8
  • PDF
A Bayesian Approach for Policy Learning from Trajectory Preference Queries
TLDR
We consider the problem of learning control policies via trajectory preference queries to an expert. Expand
  • 89
  • 7
  • PDF
...
1
2
3
4
5
...